Win Fabulous Prizes!*
Background
In my never-ending search to catalog all the web’s social networks, I have discovered a bunch of blogging sites based on the LiveJournal code. These include DeadJournal, GreatestJournal, and InsaneJournal.com to name a few. This is great for me because these sites all have built-in social networks, and all of that is produced in FOAF format.
I am current working on a project with some students to start using spidered FOAF data with a firefox extension to integrate some social network information into browsing. (We are writing a paper on this that I’ll link to just as soon as it’s finished). I’ve always maintained in my work that social networks aren’t really useful for much if we just sit there and play with them. They become powerful once we start using the data from them.
I’ve hinted at another potential use, combining social networks with tagging. But I really am interested in some other ideas. Thus, I introduce
The 2007 Unofficial FOAF Challenge
So there is a lot of social network data out there in FOAF format - easily machine readable and relatively easy to aggregate and merge. If a service were available that let you use it, what do you think would be a good way to integrate the social information into systems?
Please post your answers here. Best idea wins. Entries due by January 4, 2007. Winners will be announced January 5, 2007. Everything is totally at my sole discretion and whim.
Prizes! Well, prize singular. I hereby promise the winner their choice of a fabulous RDF, OWL, or Social Network mini button. Geek Flair.
|
|
 |
I'd use it to populate 3D worlds with avatars and situations where frequency of messaging and the density of the network enable a richer environment (more servers, more polygon budgets) depending on the numbers and kinds of successful or say, community-enhancing interactions occur.
Classic tit for tat modeling.
Now you have to ask yourself: is a company-sponsored site such as what IBM, Sun, etc., do on SecondLife a true FOAF social network? I'd say yes. Opinions will vary but shared values and shared tasks are what define and sustain communities anyway. The fact of an FOAF declaration doesn't mean they actually get together often and do anything. That is the flaw in the numbers published for those sites.
Pierre - Good idea, but it's been done. Lise Getoor, a professor at University of Maryland, College Park, has done extensive work in this space. You may be interested in her papers:
http://www.cs.umd.edu/~getoor/
A big question I want FOAF to answer is "do I know [person] in any way, via some small number of links?" This is hard right now because the data are separated into multiple silos (friendster, LJ, personal home pages with FOAF, etc). So my first app would be:
Person 1: _search for the foaf record of the first person here_
Person 2: _search for the foaf record of the second person here_
[find connections]
and the results would be like:
Drew -(friendster)-> Al -(friendster)-> Betty -(LJ)-> Charlie -(LJ)-> Jennifer
Speaking of FOAF, look for FOAF exports on Advogato in the next few days...
Sorry I'm to late for the challenge, cheers though.
pappas